I

c

I LA-UR- -7 , ' Approved for public release;

distribution is unlimited. Title:

Author(s):

Submitted to

Los Alamos N A T I O N A L L A B O R A T O R Y

Performance Estimates for Waste Treatment Pyroprocesses in ATW

Ning Li, MST-lO/LANSCE-ADTT

Second International Conference on Accelerator-Driven Transmutation Technologies and Applications

MASTER

Los Alamos National Laboratory, an affirmative actiodequal opportunity employer, is operated by the University of California for the U.S. Department of Energy under contract W-7405-ENG-36. By acceptance of this article, the publisher recognizes that the U.S. Government retains a nonexclusive, royalty-free license to publish or reproduce the published form of this contribution, or to allow others to do so, for US. Government purposes. Los Alamos National Laboratory requests that the publisher identify this article as work performed under the auspices of the U.S. Department of Energy. The Los Alarnos National Laboratory strongly supports academic freedom and a researcher's right to publish; as an institution, however, the Laboratory does not endorse the viewpoint of a publication or guarantee its technical correctness. Form 836 ( 10/96)

DISCLAIMER

This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, make any warranty, exprrss or implied, or assumes any legal liabili- ty or responsibility for the accuracy, completeness, or usefulness of any information, appa- ratus, product, or proass disdosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessar- ily state or reflect those of the United States Government or any agency thereof.

Portions of this document may be iIlegi%le in electronic image products. Images are produced from the best avaiiable original dOrlrmen.t,

Performance Estimates for Waste Treatment Pyroprocesses in ATW

Ning Li MST-lO/LANSCE-ADT7: MS H8.54

Los Alamos National Laboratov, NM 87545

Abstract: We have identified several pyrometallurgical processes for the conceptual ATW waste treatment cycle. These processes include reductive extraction, electrowinning and electrorefining, which constitute some versatile treatment cycles for liquid-metal based and molten-salt based waste forms when they are properly integrated. This paper examines the implementation of these processes and the achievable separations for some typical species. We also present a simple analysis of the processing rates limited by mass diffusion through a thin hydrodynamic boundary layer. It is shown that these processes can be realized with compact and efficient devices to meet the ATW demand for the periodic feeding and cleaning of the waste.

INTRODUCTION

Accelerator-driven Transmutation of nuclear Waste (ATW) uses a high current proton accelerator to drive an intense spallation neutron source to transmute the actinides and select fission products in a subcritical reactor-like assembly. To achieve complete burn and efficient operation on a wide variety of waste forms as a generic waste burner, an ATW system requires simple, reliable, efficient and flexible waste preparation and cleanup procedures to periodically remove the fission products[ 11.

Although aqueous separation technologies (e.g. PUREX) represent the mainstream commercial nuclear fuel processing and are efficient for their specific applications, they are ill-suited for ATW. The separation of pure plutonium presents a major proliferation risk. The various chemical reagents and processing media generate large quantities of mixed hazardous waste which exacerbates the nuclear waste problem. The separation processes need to continuously transform the waste forms back and forth for transmutation and cleanup purposes. It is also realized that aqueous ATW systems do not utilize accelerators effectively and prevent the use of advanced nuclear reactor technologies.

Realizing that waste treatment cycle selection is imperative for developing a practical ATW system, the Los Alamos ADTT Project focused on evaluating various advanced separation technologies. The basis of these technologies were explored in the Molten Salt Reactor Experiment (MSRE) at Oak Ridge (ORNL)[2], the plutonium metal work at Los Alamos (LANL)[3], and mostly notably, the Integrated Fast Reactor (IFR) program at Argonne (ANL)[4]. We found that several forms of pyrochemical processes have significant advantages over conventional aqueous technologies for ATW applications: the working mechanisms are relatively simple, the species are separated well between groups but poorly within the groups, only a few reagents are needed which are mostly used to substitute carrier components, the processing media are more radiation resistant and can be reused, and the waste streams consist of mostly pure and partitioned fission products. These processes

promise vary high equilibrium separation ratiosE3 1. We studied the separation rates of these processes under practical implementation conditions and found them sufficiently high to run ATW systems in fairly clean and neutron-efficient modes. We were thus able to integrate these pyrochemical separation technologies into simple, reliable, efficient and flexible ATW waste treatment cycles. It is noteworthy that the flow sheets thus constructed are similar for both front-end preparation and back-end cleanup, and for both molten-salt and liquid-metal waste forms, with some simplification for liquid-metal waste.

The basic processes of the ATW waste treatment cycles include reductive extraction (RedEx), electrowinning (El Win) and electrorefining (ElRef). Experiments in some molten salt systems provided the proof-of-concept[2]. Some forms of engineering implementation of these processes are partially successful[5]. The principle mechanisms are based on redox and electrochemical processes, and there is little waste due to the “ion excha~~ge”-like operation. The carrier and processing media are liquid metals (LM, bismuth in this paper) and molten salts (MS, fluorides in this paper). Efforts have been made to understand the experiments using equilibrium conditions and limiting mass transport properties[6]. This paper presents some practical conceptual implementation of these processes and estimates the limiting processing rates relevant to ATW systems.

SEPARATION MECHANISMS AND IMPLEMENTATIONS

The pyrochemical processes we selected for ATW fuels are all liquid-metal and molten-salt based. The separation is achieved by utilizing the grouping of the redox potentials for the targeted groups of species[3]. The “reagents” are used to substitute the fission products (FPs) while reconstitute the carriers or media (if properly selected). The waste streams, which consist of reduced FPs, are nearly pure, partitioned and can be disposed readily in engineered storage or permanent repository.

Reductive Extraction

Reductive extraction utilizes the difference in distributions of various species in contacting fluids to achieve separation. In the RedEx process (Fig. l), fluoride molten salt containing fission product (lanthanides) and actinide (Ac) fluorides is brought into contact with liquid bismuth containing low concentrations of lithium. Since lithium fluoride is more stable than the FP and Ac fluorides, the FPs and Acs are reduced by the lithium and become dissolved in the bismuth. Because the Ac fluorides are less stable than the FP fluorides, there are more actinides reduced into the bismuth, thus creating a separation of lanthanides and actinides in the LM and the MS. Due to the wide separation of the oxidation potentials between the groups of the most FPs (lanthanides) and Acs, the separation is high and can be enhanced via multistaging.

This separation process has been tested quite successfully in laboratory experiments[2]. The ratio of the concentration of a specific element in LM with respect to

its concentration in MS is termed the distribution coefficient under certain conditions. The ORNL experimental results of some of the distribution coefficients for one molten salts are reproduced in Fig. 2[2]. Note the large differences in the distribution coefficients for FPs and Acs.

Li-Bi

Pii clad ,ictirridcs Rich

Reductive extraction (RedExJ can be effectively used for separating elements of different molecular stability (e.g. actinides and lanthanide fission products in molten salts).

A RedEx separator: a contactor/mixer a separator

Separations: Less stable elements reduced into liquid metal

(Actinides in mi, Lanthanides mostly in MS)

cam. in LBi conc. in bfs Rafes: Distribution coeff. =

Interfacial mass transfer limited.

Fig. 1. A schematic illustration of the reductive extraction separation.

U I -

.,....,., O.'Ob.oi 0.1 1 10

Fig.2. Exp riment 1 distribution coefficients DM for select actinides and rare earths in molten salts (72%LiF- 16%BeF2- 12%ThF4)[2].

For MS based fuels, the separation proceeds in the following manner: at very small concentrations of lithium in LM, much of the Acs are reduced into the LM while most of

the FPs are left in the MS. After as few as 4-stage contacting in a counter-current multistage configuration, a high separation ratio between the FPs and Acs can be achieved (1 0,000 to 1 or more) in MS. An example of the separation ratios in a counter-current multistage RedEx contactor vs the initial Li concentration in LM is shown in Fig. 3[7] . At this point, a relatively high concentration of lithium is used in LM to reduce most of the FPs from MS, which is essentially free of Acs. These FPs in LM can be extracted via precipitation or some other means. The separated Acs in LM from previous stages can then be “back- extracted” into MS by contacting MS with BiF3 additive which is less stable than the Ac fluorides. During this operation, the Acs are extracted from MS carrier in transient but not separated among themselves, which is a significant proliferation barrier.

10-

.- 0 IO8 3 tT: IO6

.9 lo4 S

* s 10’ Q

2l IO0

I 0-2

--- Lanthanide W o n

1 .o

0.8

0.6

0.4

0.2

0.0

0.000 0.002 0.004 0.006 Initial xu

Fig. 3. Separation ratios after a 4-stage counter-current centrifugal contactor for an ATW waste[7]

The RedEx separation is not necessary for LM based fuels since ElRef can achieve the same objective (described later). It can, however, be used to clean the processing MS in ElRef of FPs for reuse.

The RedEx separation process can be implemented as any kind of liquid contactors, e.g. stirred layers of liquids. For improved processing rates and better fluid separation after the contact, we studied a multistage counter-current centrifugal contactor[7] which proves to be sufficient for ATW use. A prototype pyrocontactor has been built and tested successfully at ANL [8] with processing rates approximately the same as in our analysis.

Electrowinning

Electrowinning accomplishes separations by depositing species of the lowest oxidation potential fiom an electrolytic solution in an electrochemical cell. In ElWin process

the targeted species are electrolytically deposited on a solid cathode while a sacrificial anode is oxidized into the salt solution (Fig. 4). Elements of lower oxidation potentials deposit onto the cathode first. The rate of deposition can be controlled via external potentials, and high separation ratios can be achieved. For ATW MS waste forms, the noble metals can be extracted this way with a sacrificial anode (e.g., beryllium for LiF-BeF2 carrier salt). There is no equivalent process for LM fuels.

Electrowinning (Elwin) can be used for controlled extraction of elements with low oxidation potentials (e.g. noble metals).

An ElWin cell: an anode (can be sacrificial) a cathode (substrate)

Separations: elements with lower oxidation potential deposit out to cathode first

(Noble metals to cathode, Actinides in MS)

Rates: Voltage controlled Mass transfer (in boundary layer) limited

Fig. 4. A schematic illustration of the electrowinning process.

ElWin process can be implemented as plug-flow parallel-plate electrochemical cells. The electrodes need to be periodically replaced. ElWin constitute the basic process for the electrorefining separation process, which is to be used for LM waste forms and will be described in the following section. Understanding the electrochemistry of ElWin is the basis for using ElRef.

Electrorefining

Electrorefining is the key step to separate actinides from fission products in waste preparation and cleanup processes. It is a more sophisticated form of electrowinning in that liquid anode or cathode can be used to affect the separations by controlling the activities of the species within the electrodes (Fig. 5). Combined use of liquid and solid electrodes at different stages can achieve purification of uranium, extraction of plutonium and higher actinides, separation of fission products all in one kind of devices.

There are a variety of ways to implement ElRef. In general the anode can be solid or

liquid, one cathode has to be solid to accept uranium deposition while the other is liquid metal to accept plutonium and other actinides. The electrotransport medium is molten salts. An external current source or power supply drives the electrochemical process.

Electrorefining (ElRef) is the key step of the pyrometallurgical processing of metallic fuels for separating uranium and plutonium from each other and from fission products.

ElRef = RedEx (Electrode I MS) + ElWin (MS)

Fig. 5. A schematic illustration of an electrorefmer.

The following ElRef example is based on the ANL fuel treatment experiments and adapted for ATW waste: oxide spent fuel is decladded and crushed to increase the surface area. It is passed through a calcium reduction process in which uranium, actinides and fission products are reduced by calcium and transformed into metallic form. This metallic lump is then dissolved in liquid bismuth and used as the anode in an electrorefiner. A fluoride or chloride MS is used as the transport medium. The noble metals in the anode stays behind while U, Pu and other Acs, FPs distribute into MS. A low carbon steel rod serves as a cathode and starts to accept U deposit since uranium salt is the easiest to reduce among the heavy metals. Pu and other Acs, FPs will accumulate in MS. When U is nearly depleted and there are sufficient amount of Acs and FPs in MS, the solid cathode with U deposit is removed and the liquid bismuth anode starts to accept Acs and FPs. The difference in the oxidation potentials of Acs and FPs (rare earths) causes more Acs to go into bismuth, leading to separations. If high separation ratio is desired, multistaging can be used. The bismuth containing Acs can be used in the ATW burner without much fabrication work if metallic waste form is preferred. The MS medium will need to be cleaned when FPs accumulate to certain amount via RedEx process as described before.

The metallic waste of ATW can be directly fed into the electrorefiner for cleanup. Electrorefiners are the key elements in the ATW waste preparation and treatment facility. There is considerable development and expertise that exists both at Los Alamos and Argonne, which can be adapted for ATW use.

PERFORMANCE BASED ON LIMITING RATES

These pyroprocesses, esp. electrowinning and electrorefining, have complex electrochemical kinetics. They are also heavily dependent upon the hydrodynamic conditions of the solutions. A general description of the kinetics of these processes is beyond the scope of this paper. However, it is very useful to understand the contributing effects and the limiting steps. We can gain enough insight at this stage to show that the separation rates can be sufficient without severe demands on device and operation.

In reductive extraction, the separation mechanism is relatively simple. Our study is also facilitated greatly by the existence of experimental data and results. At the processing temperatures above 500 OC, the reaction kinetics of the oxidation and reduction at the contact interface is so fast that the process is rate-limited by intraphase mass transport to the interface. In practical implementations, the use of turbuIent mixing flow ensures that the concentrations are nearly homogeneous throughout the solution except in the boundary layers at the interface (between two contacting solutions). Then the limiting mass transfer rate is determined by mass diffusion across those boundary layers[7].

Mass diffusion also limits the throughput in electrowinning and electrorefining. Since electrowinning is the basic process among the two, we will examine it here. Parallel plate electrochemical cell has been extensively studied for its industrial applications[9]. The deposition current includes contributions from ion migration, advection, diffusion etc. For practical applications, there are important simplification one can apply to the modeling. In turbulent through flow or stirred flow between the electrodes, the species concentrations can be assumed constant throughout the bulk of the flow due to turbulent mixing. The migration and diffusion occur in the two thin boundary layers at the electrodes. The kinetics of the electrochemical processes at the electrodes can be obtained from Tafel or Butler- Volmer relations. It is found, however, that at high cell voltages, the deposition current approaches a limiting value and is determined by mass diffusion across the boundary layers [9 J. Using this limiting current in modeling electrorefiner performance achieved fair agreement with results from ANL experiments[6]. The separations depend on the process implementation and throughput in this case and will not be discussed here.

It is now possible to estimate the throughput of these pyrometallurgical processes based on the mass-diffusion-limited rates. Given an implementation these are the highest rates achievable. One can, however, choose configurations to maximize limiting rates. For simplicity, we use the configuration of a layer of turbulent mixing liquid, thickness h, on top of another liquid (as in RedEx) or solid (as in ElWin and ElRef). The bulk concentration of the target species is c, and the corresponding concentration at the interface is 0 due to reduction (or oxidation). The boundary layer thickness is 6, which can be determined from hydrodynamic conditions (Reynolds number) with scaling laws. So the mass diffusion flux

is j = D (c - 0)/6.

From mass balance d(ch)ldt = - j = - Dc/6,

the concentration evolves as c = co exp(- th) ,

where co is the initial concentration, ‘1: = (h /6) (62/0) is a diffusion time constant. To calculate the average flux, we assume the mass transfer is 95% complete, then t = 3.07, ‘

<J> = 0.95coh/(3.0 hl6 62/0) = 0.32 Dc&.

For liquids, the mass diffusion coefficient is typically D = 10-5 cm2ls. It is very easy to realize a boundary layer thickness 6 = 0.001cm. For lwt?’ solutes in 2 g/cm3 solutions (molten salts), co = 0.02 g/cm3, then <J> = 6.4~10-5 g/cm2s. If we make the interface area to be 1 m2, then the throughput is 55 kg/day. This resembles the processing rate of the ANL pyrometallurgical waste treatment devices, which is very large for a compact device. It is sufficiently high for ATW applications where even a 500 ton spent fuel per year facility only needs to process about 70 kg of actinides and fission products every day. Because of this efficiency in addition to the radiation resistance of the media, the process inventory can be very low. Multistaging only requires more devices, but the throughput remains the same.

It is easy to see that to maximize the throughput, the device needs to maximize the interface area, or minimize the boundary layer thickness. One could also wait a little less by allowing each of a multistage process to be less complete, but achieve the same separation with additional stages. It should also be noticed that large differences in mass diffusion coefficients can cause very different separation in practice than the equilibrium values. A centrifugal contactor increases the interface area by many folds because of the breaking up of one liquid into tiny droplets in the continuous phase of the other. For ElWin and ElRef electrochemical cells, vigorous stirring can decrease the diffusion boundary layer thickness. Using many small electrodes instead of large ones can increase the interface area. It should be understood, however, that these requirement can be easily fulfilled without severe demands on the engineering and material capabilities.

CONCLUSION

We have briefly described the mechanisms and performance of the key pyroprocesses for ATW waste preparation and cleanup. These processes operate on relatively simple thermodynamic principles to achieve high separation ratios, use recyclable media with few extraneous reagents, produce nearly pure and partitioned fission product waste streams. They can all be implemented in compact, reliabIe and eEcient devices with relative ease. No pure plutonium and weapons materials are present at any stages thus they are more proliferation resistant. These features of the waste treatment processes allow us to fully

explore the distinct advantages of accelerator-driven transmutation systems.

ACKNOWLEDGEMENTS

The author gratefully acknowledges the contributions by Mark Williamson, Yuchou Hu, and Francesco Venneri for the work included in this paper. This research is funded through the Los Alamos National Laboratory LDRD.

[ 11 Venneri, F., “The Physics Design of Accelerator-Driven Transmutation Systems”, in Proceedings of the International Conference on Accelerator-Driven Transmutation Technologies and Applications, Las Vegas, NV, U.S.A., 1995, pp. 117-137.

[2] Ferris, L.M., et al, “Equilibrium Distribution o f Actinide and Lanthanide Elements between Molten Fluoride Salts and Liquid Bismuth Solutions”, J. Inorg. Nucl. Chem., 32 (1969) pp. 2019-2035.

[3] Williamson, M., and Venneri, F., “Front-end and back-end electrochemistry of molten salt in accelerator- driven transmutation systems”, in Proceedings of International Conference on Evaluation of Emerging Nuclear Fuel Cycle Systems, Global 95, Versaille, France, 1995, pp.1147-1153.

[4] Burris, L., et al, “The Application of Electrorefining for Recovery and Purification of Fuel Discharged &om the Integral Fast Reactor”, AIChe Symp. Ser. 83(254)(1987) pp. 135-142.

[5] Battles, J.E., Miller, W.E., and Gay, E.C., “Pyrometallurgical processing of Integral Fast Reactor metal fuels”, in Proceedings of International Conference on Nuclear Fuel Processing and Waste Management, British Nad. Forum, Sendai, Japan, 1991, p. 342.

[6] Kobayashi, T., and Tokiwai, M., “Development of TRAIL, a simulation code for the molten salt electrorefining of spent nuclear fuel”, Journal of Alloys and Compounds 197 (1993) pp.7- 16.

[7] Li, N., et al, “Mixing o f Molten Salt in Liquid Metal in a Turbulent Centrifugal Contactor”, Los Alamos National Laboratory Technical Report, LA-UR-95-4207 (1 995).

[8] Chow, L.S., et al, “Continuous Extraction o f Molten Salt Chloride Salts with Liquid Cadmium Alloys”, in Proceedings of International Conference on Future Nuclear Systems: Emerging Fuel Cycles and Waste Disposal Options, Global 93, Seattle, WA, U.S.A., 1993, and in private communications.

[9] Lapicque, F., and Storck, A., “Modelling of a continuous parallel plate plug flow electrochemical reactor: electrowinning of copper”, J. Applied Electrochemistry 15 (1985) pp.925-935.